Open AccessReproducible Naevus Counts Using 3D Total Body Photography and Convolutional Neural Networks
Author(s) -
Brigid BetzStablein,
B D'Alessandro,
Uyen Koh,
Elsemieke I. Plasmeijer,
Monika Janda,
Scott W. Menzies,
Rainer HofmannWellenhof,
Adèle C. Green,
H. Peter Soyer
Publication year - 2021
Publication title -
dermatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.224
H-Index - 92
eISSN - 1421-9832
pISSN - 1018-8665
DOI - 10.1159/000517218
Subject(s) - kappa , medicine , gold standard (test) , nevus , dermatology , lesion , convolutional neural network , population , nuclear medicine , radiology , pathology , melanoma , artificial intelligence , mathematics , computer science , geometry , environmental health , cancer research
Background: The number of naevi on a person is the strongest risk factor for melanoma; however, naevus counting is highly variable due to lack of consistent methodology and lack of inter-rater agreement. Machine learning has been shown to be a valuable tool for image classification in dermatology. Objectives: To test whether automated, reproducible naevus counts are possible through the combination of convolutional neural networks (CNN) and three-dimensional (3D) total body imaging. Methods: Total body images from a study of naevi in the general population were used for the training (82 subjects, 57,742 lesions) and testing (10 subjects; 4,868 lesions) datasets for the development of a CNN. Lesions were labelled as naevi, or not (“non-naevi”), by a senior dermatologist as the gold standard. Performance of the CNN was assessed using sensitivity, specificity, and Cohen’s kappa, and evaluated at the lesion level and person level. Results: Lesion-level analysis comparing the automated counts to the gold standard showed a sensitivity and specificity of 79% (76–83%) and 91% (90–92%), respectively, for lesions ≥2 mm, and 84% (75–91%) and 91% (88–94%) for lesions ≥5 mm. Cohen’s kappa was 0.56 (0.53–0.59) indicating moderate agreement for naevi ≥2 mm, and substantial agreement (0.72, 0.63–0.80) for naevi ≥5 mm. For the 10 individuals in the test set, person-level agreement was assessed as categories with 70% agreement between the automated and gold standard counts. Agreement was lower in subjects with numerous seborrhoeic keratoses. Conclusion: Automated naevus counts with reasonable agreement to those of an expert clinician are possible through the combination of 3D total body photography and CNNs. Such an algorithm may provide a faster, reproducible method over the traditional in person total body naevus counts.